1

Topic: That where it is possible to esteem on Iskustvennomu to Intelligence

.. To be exact about it of now almost implementable part - deep networks. What  input for dummies. To me the one-layer trained network is now clear is as a matter of fact NxM a matrix, it train fining (changing coefficients) responsible for wrong answers. How business from the multilayer is? I read that train each layer separately, but how? What is criterion of the wrong answer of the intermediate layer?

2

Re: That where it is possible to esteem on Iskustvennomu to Intelligence

Hello, johny5, you wrote: J>. To be exact about it of now almost implementable part - deep networks. J> What  input for dummies. J> To me the one-layer trained network is now clear is as a matter of fact NxM a matrix, it train fining (changing coefficients) responsible for wrong answers. J> as business from the multilayer is? I read that train each layer separately, but how? What is criterion of the wrong answer of the intermediate layer? A question directly I will not answer, for I do not understand a subject, but I can advise people who understand. Here the blog of Mr. Levenchuka which (deep learning) popularizes this subject and advances. Here Sergey Nikolenko's blog which is engaged in mathematics of all this business (writes articles, participates in conferences etc.). At it it is possible to explain this question in details. Here its article on  which can be useful something.

3

Re: That where it is possible to esteem on Iskustvennomu to Intelligence

Hello, johny5, you wrote: J>. To be exact about it of now almost implementable part - deep networks. J> What  input for dummies. It is possible to look in a tape of very live group: https://vk.com/deeplearning There permanently any  spread.

4

Re: That where it is possible to esteem on Iskustvennomu to Intelligence

Hello, johny5, you wrote: J>. To be exact about it of now almost implementable part - deep networks. About it I do not know. J> what  input for dummies. I very much like the book: M.Tim Johns. Machine intelligence programming in applications.